ENHANCING CUSTOMER LOYALTY THROUGH MACHINE LEARNING · Prescience is a fast growing advanced...
2
The client runs a loyalty program for its UK customers with the help of a third party loyalty program provider. The program involved customers earning points on every purchase made through the client’s e-commerce portal in UK. The client then paid the cost of each point earned by customers to the loyalty program provider, which could be redeemed either at the e-commerce portal or at every redeeming partner of the loyalty program provider. The key challenge that the client faced while executing the loyalty program was to capture the incremental sales and revenue the program was bringing to the business. In addition to this, there was a need to rectify the assessment process that was influenced by strong selection biases due to the absence of pre-defined tests and control groups. The client also wanted to drive a high level of fairness and efficiency while computing the program’s contribution to the company’s sales and revenue growth. An American Multinational E-Commerce Corporation Enhances Customer Loyalty through Machine Learning !"#$%&#'%# !"#$%&'() *"# +,%&-$%% THE CHALLENGES The corporation runs a large online auction platform and shopping website. It has over 170 million active buyers globally and one billion active listings. It has a significant amount of transaction history and processes more than 100 petabytes of data daily. ENHANCING CUSTOMER LOYALTY THROUGH MACHINE LEARNING
Transcript of ENHANCING CUSTOMER LOYALTY THROUGH MACHINE LEARNING · Prescience is a fast growing advanced...
The client runs a loyalty program for its UK customers with the help of a third party loyalty program provider. The program involved customers earning points on every purchase made through the client’s e-commerce portal in UK. The client then paid the cost of each point earned by customers to the loyalty program provider, which could be redeemed either at the e-commerce portal or at every redeeming partner of the loyalty program provider.
The key challenge that the client faced while executing the loyalty program was to capture the incremental sales and revenue the program was bringing to the business. In addition to this, there was a need to rectify the assessment process that was influenced by strong selection biases due to the absence of pre-defined tests and control groups. The client also wanted to drive a high level of fairness and efficiency while computing the program’s contribution to the company’s sales and revenue growth.
An American Multinational E-Commerce Corporation Enhances Customer Loyalty through Machine Learning
!"#$%&#'%#!"#$%&'() *"# +,%&-$%%
THE CHALLENGES
The corporation runs a large online auction platform and shopping website. It has over 170 million active buyers globally and one billion active listings. It has a significant amount of transaction history and processes more than 100 petabytes of data daily.
ENHANCING CUSTOMER LOYALTY THROUGH MACHINE LEARNING
We developed a new methodology that used a cohort approach based on when customers link their online shopping portal account to the loyalty program. We matched the details with available demographics and behavioural information to minimize selection bias. We then created a classifier with known user demographics data, transaction history and program enrolment status to predict the user’s propensity of enrolling into the program.
Our approach to the challenge leveraged machine learning to determine a non-loyalty user for every loyalty user with the same propensity of consumption and enrolling into the program, and then compared the gap in their future performances.
We used the propensity score in the matching procedure to find out if each loyalty user (test user group) was matched to the non-loyalty group (control user group) or not.
!"#$%&#'%#!"#$%&'() *"# +,%&-$%%
THE SOLUTION
Prescience was able to develop a synthetic control group that helped the client in computing the program’s incremental sales and revenue generation. The model also helped in assessing how much the value of new linkages was, by comparing their GMV (Gross Merchandising Value) against the control group of a particular cohort in the pre- and post-period.
The client could also determine the program’s ROI, the incremental revenue and the monthly and quarterly running costs of the loyalty program.
THE IMPACT
Prescience is a fast growing advanced analytics company that helps enterprises become more PRESCIENT by providing meaningful business insights and recommendations generated through careful analyses of data. We leverage our tools & frameworks, deep expertise in machine learning, advanced data analytics techniques and domain knowledge along with the business knowledge of our customers to create tangible data driven solutions.
Visit us at www.prescienceds.com or send us an email at [email protected] to get in touch with us. You can also follow us on LinkedIn .
About Prescience Decision Solutions
Copyright 2019 Prescience Decision Solutions. All Rights Reserved.
